Distinguishing between Homogeneous and Heterogeneous Catalytic Activity in C-H Arylation of an Indole with Aryl Halides under ligandless Conditions: Crucial Evidence from Real Catalytic Experiments

Article abstract of DOI:10.1021/acs.oprd.9b00096

The kinetic approach for distinguishing between homogeneous and heterogeneous catalysis mechanisms by the analysis of differential selectivity was applied for direct C-H arylation of an indole with aryl halides under ligandless conditions. The differential selectivity of competing arylation of an indole with two aryl iodides or bromides and the differential regioselectivity of a parallel formation of C2- and C3-regioisomers of arylated indoles were studied by the simple method of phase trajectories needed for raw kinetic data only. The results obtained indicated that the selectivities strongly depended on the type (soluble or insoluble) and concentration of the Pd precursor. Considering the pathways of Pd transformations under the reaction conditions, this indicates a substantial contribution of catalysis occurring on the surface of metallic Pd (including nanoparticles).

Full text of DOI:10.1021/acs.oprd.9b00096

Article
pubs.acs.org/OPRD
Cite This: Org. Process Res. Dev. XXXX, XXX, XXX−XXX
Distinguishing between Homogeneous and Heterogeneous
Catalytic Activity in C−H Arylation of an Indole with Aryl Halides
under “Ligandless” Conditions: Crucial Evidence from Real Catalytic
Experiments
Elena V. Yarosh, Anna A. Kurokhtina, Elizaveta V. Larina, Nadezhda A. Lagoda,
and Alexander F. Schmidt*
Department of Chemistry, Irkutsk State University, 1 K. Marx Str., Irkutsk 664003, Russia
S
* Supporting Information
ABSTRACT: The kinetic approach for distinguishing between homogeneous and heterogeneous catalysis mechanisms by the
analysis of differential selectivity was applied for direct C−H arylation of an indole with aryl halides under “ligandless”
conditions. The differential selectivity of competing arylation of an indole with two aryl iodides or bromides and the differential
regioselectivity of a parallel formation of C2- and C3-regioisomers of arylated indoles were studied by the simple method of
phase trajectories needed for raw kinetic data only. The results obtained indicated that the selectivities strongly depended on the
type (soluble or insoluble) and concentration of the Pd precursor. Considering the pathways of Pd transformations under the
reaction conditions, this indicates a substantial contribution of catalysis occurring on the surface of metallic Pd (including
nanoparticles).
KEYWORDS: C−H arylation, palladium, differential selectivity, phase trajectories, heterogeneous catalysis
operando ones) are ineffective for obtaining reliable results
for this reaction. In order to draw any valid conclusions about
the nature of the active species, it is necessary to apply an
approach leading to results that are independent of the
concentration of the active species, which varies in the above-
mentioned processes. In addition, using any model conditions
(i.e., low substrate/catalyst ratio or excluding any components
of a reaction system) must be avoided due to crucial influence
of all reaction components to each other and, in the first place,
to catalyst transformations.2−4 We previously proposed the
kinetic study of the differential selectivity of competing
reactions because it does not depend on the concentration
of the active species and is determined by their nature alone.5
Therefore, a change in the differential selectivity under varying
reaction conditions unambiguously points to the changes in
the species responsible for catalysis. As a convenient method
for the qualitative and quantitative estimation of differential
selectivity, the use of the so-called phase trajectories of the
reactions was proposed.4−6 To the best of our knowledge, the
phase trajectories have not been used as the instrument for
mechanistic studies of catalytic reactions earlier; however, they
are widely used for the analysis of the dynamical system
behavior in physics and engineering. With regard to the
catalytic reaction, if two competing substrates are used, the
phase trajectory is the interdependence of the concentrations
of the products formed from the competing substrates. In this
case, the slope of the tangent to any point of the phase
trajectory is the ratio of the rates of the competing reactions,
INTRODUCTION
■
Arylation of indoles via direct C−H activation (direct
arylation) is a very attractive synthetic approach for obtaining
up-to-date pharmaceuticals, agrochemicals, etc.1 Nevertheless,
a crucial mechanistic question on the truly homogeneous or
heterogeneous nature of catalysis in the reaction has not been
precisely identified. However, while the question is not solved,
it is impossible not only to develop a rational approach to the
elaboration of new catalysts for direct C−H activation methods
and to optimize the synthetic protocol under laboratory and
technological conditions but also to construct more reliable full
mathematical (macrokinetic) models of processes and
economical reactor design.
Studies aimed at distinguishing these two mechanisms have
been conducted (for a review of such papers, see refs 2 and 3
and a recent review on the related cross-coupling processes4);
however, the different reaction tests for homogeneity/
heterogeneity as well as different ex situ techniques for
detection of a Pd form used in these studies are unable to solve
the problem unambiguously. The matter is that the reaction is
complicated by the behavior of several consecutive parallel
processes including the interconversions of soluble and
insoluble Pd forms occurring simultaneously with the catalytic
reaction. So, the results of the tradition reaction tests for
homogeneity/heterogeneity were subjected to the influence of
such processes. Pd dissolution from the surface, as well as its
reverse precipitation, leading to the formation of nanoparticles
and bulk metal, have been well-established using various
methods under the reaction conditions and occur undoubtedly,
at the very least, in “ligandless” conditions.2,3 Therefore, any
conventional research methods (including kinetic and
Received: February 26, 2019
© XXXX American Chemical Society
A
DOI: 10.1021/acs.oprd.9b00096
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
which definitively characterizes the differential selectivity.5
Therefore, with the coinciding/differing phase trajectories
obtained by varying the reaction conditions, it is possible to
make valid conclusions about the coinciding/differing natures
of the active species formed in situ during the reaction.
Recently we have published the paper dealing with the
investigation of differential selectivity of indole direct arylation
(with two competing aryl halides, two indoles, and non-
competing variant analyzing differential regioselectivity of C2/
C3-phenyl indole formation) using the phase trajectories.7
With the natures of the cation and anion of the base as well as
of inorganic salt (using as additive to the catalytic system)
varied, the anionic type of active species has been established.
Also, the data obtained allowed concluding of the participation
of indole in the electrophilic substitution, at least under
“ligandless” conditions. However, the obtained results on the
types of elementary steps realized in the catalytic cycle of
indole direct arylation and about anionic type of active species
did not provide any information about realization of catalysis
in the solution phase or on the surface of heterogeneous Pd
species. Therefore, to distinguish these two possibilities, an
extra study needs to be carried out.
It is well-known that the type (soluble or insoluble) as well
as concentration of the Pd precursor are the crucial factors
influencing the Pd distribution between the soluble and
insoluble forms during the catalytic reaction. Additionally, the
type and concentration of the precursor also determine the size
and/or shape of the Pd metal particles formed in the solution
and/or on the heterogeneous support.2,3,8 It is very important
that when using “ligandless” reaction conditions (the absence
of any strong organic ligands), the Pd molecular complexes in
solution remain unchanged when the nature or concentration
of the Pd precursor varies.9,10 Therefore, the invariability of the
phase trajectories (i.e., of differential selectivity) using soluble
and insoluble (for instance, metallic Pd deposited on various
supports) Pd precursors indicates unambiguously that the
nature of the active species in these reactions is the same and is
independent of the precatalyst nature. This rules out the
activity of the supported palladium because it is absent in the
reaction using homogeneous precatalysts.4−6 Furthermore, any
changes in the differential selectivity when moving from a
soluble precursor to a supported one become possible only
when heterogeneous catalysis occurs. Therefore, the differing
phase trajectories obtained by varying the type (soluble or
insoluble) and/or concentration of catalyst precursors allow
valid conclusions about the considerable contribution of the
heterogeneous catalysis mechanism in the formation of the
reaction products. It is important to emphasize that possible
contributions of homogeneous and heterogeneous mechanisms
into observed catalytic activity are likely to vary substantially
under the reaction proceeding due to aggregation of Pd and its
inverse solubilization. However, the resulting phase trajectories
of the reactions differing in such contributions would also be
different, as it is under the sole heterogeneous catalytic activity.
So, any discrepancy of the phase trajectories observed with
varying the factors mentioned above definitely points to the
direct participation of heterogeneous Pd in catalysis.
performed under real catalytic conditions exclude any possible
restrictions for the conclusions to be used for real catalytic
ones.
EXPERIMENTAL SECTION
■
General Considerations. All reactants and solvents were
obtained from Sigma-Aldrich or Acros (grade p.a.) and were
used as received without further purification or drying. The
qualitative and quantitative analysis of the C2- and C3-arylated
products was performed by gas chromatography (GC,
Chromatec Crystal 5000.2 instrument fitted with a flame
ionization detector [FID] and a 15 m HP-5 methyl phenyl
siloxane capillary column) or GC−MS (Shimadzu GC−MS
QP-2010 Ultra, ionization energy of 70 eV, 0.25 μm × 0.25
mm × 30 m GsBP-5MS column, with He as the carrier gas).
The recorded mass spectra were compared with those available
in the literature (Wiley, NIST, and NIST05 comparison
libraries).
Samples for GC and GC−MS analyses were collected at
different reaction time points. Concentrations of C2- and C3-
arylated products were determined by GC−MS based on the
relative area of GC−MS signals referred to an internal standard
(naphthalene) calibrated to the corresponding pure com-
pound. To estimate the reproducibility of the data, each
experiment was performed three times. For phase trajectories
plotted, the appropriate polynomial fitting of the experimental
data was used in order to be convinced in phase trajectories
overlapping/changing.
Catalyst Preparation. Pd/C was prepared using Pd-
(OAc)2 at a Pd loading of 4 wt %. The support material
(“sibunit”, 1 g) was suspended in 20 mL of toluene. Pd(OAc)2
was added, and the suspension was stirred for 30 min at 95 °C.
Completeness of Pd(OAc)2 adsorption was controlled by the
observation of changes in intensity of the absorption band in
solution at 300 nm. Then 0.04 mL of formic acid was added to
reduce palladium on the carbon surface, and the suspension
was stirred for 20 min until discoloration occurred. The
resulting catalyst was filtered and washed with acetone. The
catalysts were used after drying in a vacuum.
The procedure for Pd/Al2O3 synthesis was analogous to
those for Pd/C. Pd black was obtained by the reduction of a
2.5 M solution of Pd(acac)2 in DMF by an equimolar amount
of NaBH4 under Ar.11 The completeness of Pd(acac)2
reduction was controlled by the observation of changes in
the intensity of the absorption band in solution at 330 nm.
The resulting catalyst was precipitated by hexane and washed
by ethanol. The catalysts were dried in a vacuum and stored
under Ar.
Competing Direct Arylation of Indole by Two Aryl
Halides. Two competing aryl halides (1.25 mmol each),
indole (2.5 mmol), base (NaOAc, if another is not specified,
1.625 mmol), naphthalene (0.5 mmol, as an internal standard
for GC and CG−MS analyses), and 1.6 mol % Pd precursor
(0.04 mmol) were introduced into a glass reactor equipped
with a magnetic stir bar and a septum inlet. Five mL of a DMF
and water mixture (4/1) was added; the mixture was placed in
a preheated oil bath (140 °C). The reaction was performed
with vigorous stirring at the specific temperature for 4−6 h.
The following Pd compounds were used in order to
investigate the influence of the catalyst precursor nature on
the differential selectivity: PdCl2, Pd/C, Pd/Al2O3, Pd black.
The following concentrations of PdCl2 were used in order to
investigate the influence of the catalyst precursor concentration
It should be particularly noted that such a rapid and efficient
approach for distinguishing between the homogeneous or
heterogeneous nature of active species needs the concentration
of organic reaction products data only (raw GC or GC−MS
data), which makes such conclusive methodology become very
attractive for kinetic investigations. Moreover, such studies
B
DOI: 10.1021/acs.oprd.9b00096
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
on the differential selectivity: 0.008 mmol (0.32 mol %), 0.02
mmol (0.8 mol %), 0.04 mmol (1.6 mol %), 0.08 mmol (3.2
mol %). The experiment with HCOONa or NaBH4 (0.5
mmol) and with NBu4Br (0.8 mmol) was performed using 1.6
mol % of PdCl2.
Noncompeting Direct Arylation of Indole by Iodo-
benzene. Indole (2.5 mmol), iodobenzene (2.5 mmol), base
(1.625 mmol), naphthalene (0.5 mmol, as an internal standard
for GC and CG−MS analyses), and 1.6 mol % Pd precursor
(0.04 mmol) were introduced into a glass reactor equipped
with a magnetic stir bar and a septum inlet. Five mL of a DMF
and water mixture (4/1) was added; the mixture was placed in
a preheated oil bath (140 °C). The reaction was performed
with vigorous stirring at the specific temperature for 4−6 h.
The following Pd compounds were used in order to
investigate the influence of the catalyst precursor nature on
the differential selectivity: PdCl2, Pd/C, Pd/Al2O3, Pd black.
the competing conditions (see Figure 2b demonstrating the
similarity of regioselectivities obtained under competing and
noncompeting conditions). The latter result indicates the
absence of any mutual influence of the additional aryl halide
(used under competing conditions) on the formation of the
active species responsible for the parallel formation of the C2-
and C3-phenylindoles.
The second important feature of the phase trajectories
plotted by using the concentrations of the C2- and C3-
regioisomers was the clearly defined nonlinearity of some of
them, particularly, using PdCl2 (Figures 1c and 2a). When the
parallel formation of two products from one common
intermediate of a catalytic cycle or from a common
intermediate and a common reagent (that corresponds to
regioisomers formation) occurs, strictly linear phase trajecto-
ries should, in theory, be expected (constant slope of the
tangent to any point of the phase trajectory),5 as opposed to
the formation of two products from one common intermediate
of a catalytic cycle and two different competing reagents (sum
of C2- and C3-arylated products from competing aryl halides).
In the most common case, the ratio of the rates of product
formation in competing reactions, which corresponds to the
slope of the phase trajectory, consists of the ratio of the rate
constants of the definite steps of these reactions and the ratio
of the competing substrates concentrations varying in the
course of the reaction. If the parallel (noncompeting)
formation of two regioisomers occurs, the competing
substrates are absent, and the common reagent (indole, in
this case) acts as “competing substrates” reacting with the
common intermediate of the parallel catalytic cycles. As a
result, the ratio of the substrate concentrations is canceled in
the ratio of the rates of the regioisomers formation.5 Therefore,
the rate ratio becomes equal to the ratio of the rate constants
only, which should be invariable during the reaction if the
nature of the active species does not change. Consequently, the
phase trajectory plotted using the concentrations of the parallel
(noncompetingly) forming products of the reaction should be
strictly linear. Therefore, nonlinearity of the phase trajectories
for the parallel formation of regioisomers (Figures 1c and 2a)
is direct evidence for the change in active species during the
catalytic reaction,7 and according to the above text, at least one
species should be heterogeneous. The size and/or shape of the
heterogeneous Pd forms depend strongly on the concentration
of the Pd complexes in solution due to the nonlinear
agglomeration kinetics.12,13 Therefore, the differential selectiv-
ity of the competing aryl iodides was also investigated by
varying the concentration of the soluble precursor. It follows
from the data obtained that the selectivity was virtually
insensitive to the changes in the concentration (Figure 3a).
However, the coincidence of the phase trajectories should be
interpreted with caution (in contrast to their difference, which
points unambiguously to the changes in the differential
selectivity). The constancy of the phase trajectories can result
from both the differential selectivity invariance and its slight
changes along with the simultaneously insufficient sensitivity of
the analytical method used. In such situations, the maximum
possible factors capable of influencing the type of active species
should be varied. As we analyzed the regioselectivity of the C2-
and C3- phenylindoles, its slight but reproducible change with
a varying concentration of the soluble precursor can be
considered as evidence for heterogeneous catalysis (Figure 3b).
In order to widen the set of factors capable of influencing the
amount, size, and/or shape of the insoluble Pd species, we
RESULTS AND DISCUSSION
■
In order to measure the differential selectivity of competing
aryl halides, the phase trajectories should be plotted using the
sums of C2- and C3-arylated indoles forming from each
competing aryl halide. When competing arylation of indole by
iodobenzene and 4-iodoanisole (Scheme 1) was performed
Scheme 1. Direct Arylation of Indole by Competing Aryl
Iodides or Aryl Bromides
using soluble (PdCl2) and several insoluble (including those
deposited on heterogeneous supports) Pd precursors, the
phase trajectories were different (Figure 1a). As indole
arylation accompanied by the parallel formation of C2- and
C3-arylated regioisomers of the reaction products, the phase
trajectories were also plotted using the concentrations of the
C2- and C3-phenylindoles formed from iodobenzene to
measure the differential regioselectivity (Figure 1c).
According to the data obtained, the regioselectivity of the
products formed form one aryl halide and indole was also
dependent on the type of precursor used. Such result
unambiguously points to heterogeneous nature of active
species or, at least, their significant contribution to catalytic
activity along with homogeneous ones. As a single phase
trajectory for the experiments with homogeneous and
heterogeneous precatalysts would rule out supposition about
activity of supported palladium (because it is absent with
homogeneous precatalysts),4 the sole explanation for different
phase trajectories in such situation is participation of
heterogeneous Pd in catalysis. It should be noted that the
regioselectivities of the C2- and C3-phenylindoles obtained
under noncompeting conditions (i.e., the reaction where only
iodobenzene and indole were present, as shown in Scheme 2)
were also dependent on the type of precursor used (Figure 2a).
Also the regioselectivities were similar to those obtained under
C
DOI: 10.1021/acs.oprd.9b00096
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 1. Phase trajectories of the reaction of indole with competing iodobenzene and 4-iodoanisole (Scheme 1) plotted by using (a, b) sum
concentrations of the corresponding C2- and C3-arylated products and (c, d) concentrations of C2- and C3-phenylindoles, under varying the
⧫
◊
■
▲
●
nature of the Pd precursor: PdCl2 ( ), Pd/C ( ), Pd/Al2O3 ( ), Pd black (Δ), PdCl2 + HCOONa ( ), PdCl2 + NaBH4( ).
8%. At the same time, the rates of arylated indoles
accumulation decreased twice. A similar tendency of an
increase in the rates of formation of arenes and a decrease in
the rates of formation of the arylated indoles was observed
using NaBH4 as a reductant. However, as the value of
differential selectivity (i.e., slope of a phase trajectory) is
determined only by the ratio of the consumption rates of two
substrates (or two intermediates) competing for a common
intermediate (or reagent) in the particular catalytic step, any
side transformations of reactants proceeding at any another
step of the catalytic cycle (or outside it at all) cannot affect the
differential selectivity of the step. Therefore, any changes of the
observed differential selectivity (i.e., discrepancy of the phase
trajectories) are resulted from a change in the selectivity-
determining step and not from any other changes in the
reaction system. So, the differential selectivity changes
observed with both the addition of the reductant and various
natures and concentrations of the precursor used unambigu-
ously pointed to the change in active species associated with
the considerable contribution of the heterogeneous catalysis
mechanism (including catalysis by nanoparticles).
Scheme 2. Noncompeting Direct Arylation of Indole by
Iodobenzene
compared the phase trajectories obtained using the soluble
precursor PdCl2 with and without the addition of sodium
formate, which is a typical reductant used in aryl halide
coupling reactions.12 It is well-known that an additive of a
reductant influences the concentration of Pd(0) complexes in
solution, and therefore, the amount, size, and shape of the Pd
metal particles formed as a result of Pd(0) complexes
agglomeration. The differential selectivity of competing aryl
halides as well as the differential regioselectivity of the C2- and
C3-arylated indoles formed from one aryl halide were found to
be affected by the HCOONa as well as NaBH4 addition (see
Figure 1b, d). It should be noted that the concentration of a
reductant added also affected the differential selectivity and
regioselectivity. It is worth noting separately that the addition
of HCOONa led to a slight increase in the amount of benzene
formed in the side reaction of iodobenzene reduction from 5 to
It can be assumed that the relative contributions of
homogeneous and heterogeneous catalysis can considerably
vary when aryl bromides having a different reactivity (and as a
result, a different ability to retain Pd in solution as well as to
D
DOI: 10.1021/acs.oprd.9b00096
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 2. Phase trajectories (a) of the reaction of noncompeting direct arylation of indole with iodobenzene (Scheme 2) plotted by using
⧫
◊
■
concentrations of C2- and C3-phenylindoles under varying the nature of the Pd precursor: PdCl2 ( ), Pd/C ( ), Pd/Al2O3 ( ), and Pd black (Δ)
■
and (b) of the reaction of noncompeting direct arylation of indole with iodobenzene (Scheme 2) ( ) and the reaction of competing direct
arylation of indole with iodobenzene and 4-iodoanisole (Scheme 1) ( ) using Pd/C plotted by using concentrations of C2- and C3-phenylindoles
□
shown for comparison.
Figure 3. Phase trajectories of the reaction of indole with competing iodobenzene and 4-iodoanisole (Scheme 1) plotted by using (a) sum
concentrations of the corresponding C2- and C3-arylated products and (b) concentrations of C2- and C3-phenylindoles, under varying the
⧫
◊
●
concentration of PdCl2: 3.2 mol % (Δ), 1.6 mol % ( ), 0.8 mol % ( ), and 0.32 mol % ( ).
Figure 4. Phase trajectories of the reaction of indole with competing bromobenzene and 4-bromoanisole (Scheme 1) plotted by using (a) sum
concentrations of the corresponding C2- and C3-arylated products and (b) concentrations of C2- and C3-phenylindoles, under varying the
⧫
◊
●
concentration of PdCl2: 3.2 mol % (Δ), 1.6 mol % ( ), 0.8 mol % ( ), and 0.32 mol % ( ).
enable dissolution of metallic Pd) are used instead of aryl
iodides. When we carried out indole arylation with competing
bromobenzene and 4-bromoanisole under “ligandless” con-
ditions (Scheme 1), the substrates conversions using
heterogeneous precursors were insufficient to use the data
obtained for formulating any decisive conclusions on the
differential selectivity patterns. Therefore, experiments with a
varying concentration of the soluble catalyst precursor PdCl2
were performed. The differential selectivity of the competing
aryl bromides (Figure 4a) as well as the regioselectivity of the
C2- and C3-phenylindoles (Figure 4b) were sensitive to the
concentration of Pd, indicating the considerable contribution
of heterogeneous catalysis under these reaction conditions as
well.
In order to validate the conclusion, additional experiments
using another pair of competing aryl bromides, i.e.,
E
DOI: 10.1021/acs.oprd.9b00096
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Figure 5. Phase trajectories of the reaction of indole with competing bromobenzene and 4-bromochlorobenzene (Scheme 1) plotted by using (a)
sum concentrations of the corresponding C2- and C3-arylated products and (b) concentrations of C2- and C3-phenylindoles, under varying the
⧫
◊
○
●
concentration of PdCl2: 3.2 mol % (Δ), 1.6 mol % ( ), 0.32 mol %( ), and using additives of HCOONa ( ) and NBu4Br ( ).
Figure 6. Phase trajectories of the reaction of indole with competing bromobenzene and 4-bromochlorobenzene (Scheme 1) with Na2CO3 as a
base, plotted by using (a) sum concentrations of the corresponding C2- and C3-arylated products and (b) concentrations of C2- and C3-
⧫
▲
○
phenylindoles, under varying the concentration of PdCl2: 3.2 mol % ( ), 1.6 mol % ( ), 0.32 mol % ( ).
bromobenzene and 4-bromochlorobenzene, were carried out.
Under these conditions, varying the homogeneous precursor
concentration did not lead to visual changes in the differential
selectivity of the competing aryl bromides (Figure 5a).
However, the phase trajectories obtained using soluble PdCl2
with and without the addition of sodium formate were
different, indicating the participation of heterogeneous
palladium forms in catalysis for these substrates as well (see
inset in Figure 5a). As additional evidence for the reaction
proceeded through heterogeneous catalysis, the obvious
sensitivity of the differential regioselectivity of the C2- and
C3-phenylindoles in these experiments to varying the
precursor concentration and the addition of HCOONa should
be considered (Figure 5b and inset). Furthermore, the
regioselectivity changed when the additive NBu4Br was used
(Figure 5b). It is well-known that such compounds play dual
roles in the catalytic systems of aryl halides coupling; i.e., they
make both soluble Pd molecular complexes and nanoparticles
more resistant to agglomeration (i.e., act as nanoparticle
stabilizers) and facilitate the dissolution of Pd metal
particles.9,12 Since bromide anions are present in the system
even when the NBu4Br additive was not added (i.e., they are
endogenous anions formed as a result of aryl bromide
conversion), the composition of molecular complexes should
not be affected by such additive. In contrast, it is likely that the
Pd nanoparticle size and/or shape are sensitive to the presence
of a stabilizer; therefore, the regioselectivity changes when
using such additive (Figure 5b) indicate catalysis proceeding
on the surface of such particles.
Considering all of the phase trajectories obtained using the
C2- and C3-phenylindoles (i.e, the differential regioselectivity)
under both competing and noncompeting conditions, it should
be kept in mind that all of the experiments described above
were performed using NaOAc as the base. However, it was
demonstrated previously that the regioselectivity of the C2-
and C3-phenylindoles is strongly dependent on the nature of
the base used.7,14,15 So, in order to validate the conclusions
drawn using NaOAc, we carried out the experiments aimed at
distinguishing between homogeneous and heterogeneous
catalysis mechanisms under competition of bromobenzene
and 4-bromochlorobenzene using Na2CO3 as the base. As it
follows from the data obtained, varying the PdCl2 concen-
tration did not lead to considerable changes in the differential
selectivity of the competing aryl bromides, at least, at the
F
DOI: 10.1021/acs.oprd.9b00096
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Author Contributions
reaction beginning (Figure 6a), analogously to the situation
when NaOAc was used as the base. However, differential
regioselectivity of C2- and C3-phenylindoles in these experi-
ments changed (Figure 6b) pointing to catalysis proceeding on
heterogeneous palladium forms, as it was for the experiments
using NaOAc. Therefore, it can be concluded that participation
of heterogeneous Pd occurs in the indole direct C−H arylation
irrespectively on the nature of a base used.
It is worth noting that the results obtained with both aryl
iodides and bromides cannot ultimately disprove the partial
progress of the reaction in solution through the homogeneous
pathway along with the heterogeneous one. Indeed, the
nonlinearity of the phase trajectories plotted using the
concentrations of the C2- and C3-phenylindoles can result
from changes during the reaction, owing to the relative
catalytic contributions of a few (two, as a minimum) distinct
active species possessing different reaction selectivities. Some
of these species may be homogeneous. It is especially topical to
take into account the results of the related cross-coupling
reactions, which demonstrate catalysis by multiple-type active
species using different catalyst precursors.8,16−19 However, the
described results unambiguously indicate a considerable
contribution of heterogeneous catalytic activity being able to
determine the observed differential selectivity of the reaction.
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Russian Foundation for Basic
Research (16-29-10731_ofi_m).
REFERENCES
■
(1) Rossi, R.; Lessi, M.; Manzini, C.; Marianetti, G.; Bellina, F.
Direct (Hetero)arylation Reactions of (Hetero)arenes as Tools for
the Step- and Atom-Economical Synthesis of Biologically Active
Unnatural Compounds Including Pharmaceutical Targets. Synthesis
2016, 48, 3821−3862.
(2) Reay, A. J.; Fairlamb, I. J. S. Catalytic C-H bond
functionalisation chemistry: the case for quasi-heterogeneous
catalysis. Chem. Commun. 2015, 51, 16289−16307.
(3) Cano, R.; Schmidt, A. F.; McGlacken, G. P. Direct arylation and
heterogeneous catalysis; Ever the twain shall meet. Chem. Sci. 2015, 6,
5338−5346.
(4) Biffis, A.; Centomo, P.; Del Zotto, A.; Zecca, M. Pd Metal
Catalysts for Cross-Couplings and Related Reactions in the 21st
Century: A Critical Review. Chem. Rev. 2018, 118, 2249−2295.
(5) Schmidt, A. F.; Kurokhtina, A. A.; Larina, E. V. Differential
selectivity measurements and competitive reaction methods as
effective means for mechanistic studies of complex catalytic reactions.
Catal. Sci. Technol. 2014, 4, 3439−3457.
(6) Schmidt, A.; Kurokhtina, A.; Larina, E. Simple kinetic method
for distinguishing between homogeneous and heterogeneous
mechanisms of catalysis, illustrated by the example of “ligand-free”
Suzuki and Heck reactions of aryl iodides and aryl bromides. Kinet.
Catal. 2012, 53, 84−90.
(7) Kurokhtina, A. A.; Larina, E. V.; Yarosh, E. V.; Lagoda, N. A.;
Schmidt, A. F. Mechanistic Study of Direct Arylation of Indole Using
Differential Selectivity Measurements: Shedding Light on the Active
Species and Revealing the Key Role of Electrophilic Substitution in
the Catalytic Cycle. Organometallics 2018, 37, 2054−2063.
(8) Ananikov, V. P.; Beletskaya, I. P. Toward the Ideal Catalyst:
From Atomic Centers to a ‘Cocktail’ of Catalysts. Organometallics
2012, 31, 1595−1604.
CONCLUSION
■
The convenient method of distinguishing between homoge-
neous and heterogeneous catalysis by the measurement of
differential selectivity using a pair of competing substrates was
applied for direct C−H arylation of indole. Patterns of
differential selectivity using competing aryl iodides or bromides
as well as of differential regioselectivity of C2- and C3-
phenylindoles were studied under varying the factors affecting
the heterogeneous Pd species formed under the reaction
conditions. The kinetic data obtained under real catalytic
conditions unambiguously point to the considerable contribu-
tion of catalysis on the surface of the heterogeneous Pd forms
(including nanoparticles generated in situ). Because the
nonlinearity of the phase trajectories describing the regiose-
lectivity of the C2- and C3-phenylindoles clearly indicates the
participation of more than one distinct active species in
catalysis, at least one heterogeneous active species is
contributing considerably to catalysis. The data obtained are
urgent from a practical view also because it allows for
estimating the possibility of the development of a truly
heterogeneous catalytic system for the industrial application of
the process under consideration including pharmaceutical
synthesis.
(9) Kohler, K.; Kleist, W.; Prockl, S. S. Genesis of coordinatively
unsaturated palladium complexes dissolved from solid precursors
during Heck coupling reactions and their role as catalytically active
species. Inorg. Chem. 2007, 46, 1876−1883.
(10) de Vries, J. G. A unifying mechanism for all high-temperature
Heck reactions. The role of palladium colloids and anionic species.
Dalton Trans. 2006, 421−429.
(11) Titova, Y. Y.; Belykh, L. B.; Belonogova, L. N.; Shmidt, F. K.
Formation and properties of hydrogenation catalysts based on
palladium bisacetylacetonate and sodium tetrahydroborate. Russ. J.
Appl. Chem. 2009, 82, 1978−1985.
(12) Schmidt, A. F.; Al Halaiqa, A.; Smirnov, V. V. Interplays
between reactions within and without the catalytic cycle of the Heck
reaction as a clue to the optimization of the synthetic protocol. Synlett
2006, 18, 2861−2878.
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.oprd.9b00096.
Raw GC−MS kinetic data (PDF)
(13) Beck, A.; Horvath, A.; Szucs, A.; Schay, Z.; Horvath, Z. E.;
Zsoldos, Z.; Dekany, I.; Guczi, L. Pd nanoparticles prepared by
“controlled colloidal synthesis” in solid/liquid interfacial layer on
silica. I. Particle size regulation by reduction time. Catal. Lett. 2000,
65, 33−42.
(14) Joucla, L.; Batail, N.; Djakovitch, L. On Water” Direct and Site-
Selective Pd-Catalysed C-H Arylation of (NH)-Indoles. Adv. Synth.
Catal. 2010, 352, 2929−2936.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: aschmidt@chem.isu.ru.
ORCID
Alexander F. Schmidt: 0000-0002-1881-7620
G
DOI: 10.1021/acs.oprd.9b00096
Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(15) Lane, B. S.; Brown, M. A.; Sames, D. Direct Palladium-
Catalyzed C-2 and C-3 Arylation of Indoles: A Mechanistic Rationale
for Regioselectivity. J. Am. Chem. Soc. 2005, 127, 8050−8057.
(16) Kurokhtina, A. A.; Schmidt, A. F. Suzuki reaction: Mechanistic
multiplicity versus exclusive homogeneous or exclusive heterogeneous
catalysis. Arkivoc. 2009, 11, 185−203.
(17) Trilla, M.; Pleixats, R.; Man, M. W. C.; Bied, C.; Moreau, J. J. E.
Hybrid Organic-Inorganic Materials from Di-(2-pyridyl)methylamine-
Palladium Dichloride Complex as Recoverable Catalysts for Suzuki,
Heck and Sonogashira Reactions. Adv. Synth. Catal. 2008, 350, 577−
590.
(18) Eremin, D. B.; Ananikov, V. P. Understanding Active Species in
Catalytic Transformations: from Molecular Catalysis to Nano-
particles, Leaching, ‘Cocktails’ of Catalysts and Dynamic Systems.
Coord. Chem. Rev. 2017, 346, 2−19.
(19) Barreiro, E. M.; Hao, Z.; Adrio, L. A.; van Ommen, J. R.;
Hellgardt, K.; Hii, K. K. M. Spatial, temporal and quantitative
assessment of catalyst leaching in continuous flow. Catal. Today 2018,
308, 64−70.
H
DOI: 10.1021/acs.oprd.9b00096
Org. Process Res. Dev. XXXX, XXX, XXX−XXX